DiffCkt: A Diffusion Model-Based Hybrid Neural Network Framework for Automatic Transistor-Level Generation of Analog Circuits

2025-07-01

AI Summary

Pre-layout analog circuit design heavily relies on expert intuition, hindering performance-driven automation. To address this, we propose DiffCkt—the first end-to-end, transistor-level circuit generation framework based on diffusion models. DiffCkt jointly models circuit topology and device parameter distributions, integrating performance-constrained optimization with topology synthesis to directly generate implementable circuits satisfying user-specified metrics (e.g., gain, bandwidth, power). To balance generation quality and efficiency, we introduce the Circuit Generation Efficiency Index (CGEI) as a unified evaluation metric. Experiments across diverse analog circuits—including OTAs, LDOs, and amplifiers—demonstrate that DiffCkt achieves state-of-the-art performance: CGEI improves by 2.21×–8365× over prior methods, and all generated circuits meet tape-out-ready performance specifications. The associated dataset and code are publicly released.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Web Mining and Content Analysis: Web data generation and simulation
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Analog circuit design consists of the pre-layout and layout phases. Among them, the pre-layout phase directly decides the final circuit performance, but heavily depends on experienced engineers to do manual design according to specific application scenarios. To overcome these challenges and automate the analog circuit pre-layout design phase, we introduce DiffCkt: a diffusion model-based hybrid neural network framework for the automatic transistor-level generation of analog circuits, which can directly generate corresponding circuit structures and device parameters tailored to specific performance requirements. To more accurately quantify the efficiency of circuits generated by DiffCkt, we introduce the Circuit Generation Efficiency Index (CGEI), which is determined by both the figure of merit (FOM) of a single generated circuit and the time consumed. Compared with relative research, DiffCkt has improved CGEI by a factor of $2.21 sim 8365 imes$, reaching a state-of-the-art (SOTA) level. In conclusion, this work shows that the diffusion model has the remarkable ability to learn and generate analog circuit structures and device parameters, providing a revolutionary method for automating the pre-layout design of analog circuits. The circuit dataset will be open source, its preview version is available at https://github.com/CjLiu-NJU/DiffCkt.

Problem

• Automates analog circuit pre-layout design phase
• Generates circuit structures and device parameters
• Improves Circuit Generation Efficiency Index (CGEI)

Innovation

• Diffusion model-based hybrid neural network
• Automatic transistor-level circuit generation
• Circuit Generation Efficiency Index (CGEI)